神戸大学附属図書館デジタルアーカイブ
入力補助
English
カテゴリ
学内刊行物
ランキング
アクセスランキング
ダウンロードランキング
https://doi.org/10.24546/81006715
このアイテムのアクセス数:
6
件
(
2024-04-25
23:43 集計
)
閲覧可能ファイル
ファイル
フォーマット
サイズ
閲覧回数
説明
81006715 (fulltext)
pdf
7.84 MB
9
メタデータ
ファイル出力
メタデータID
81006715
アクセス権
open access
出版タイプ
Version of Record
タイトル
A Comparison of Phrase Structures in Learner and Native English Writing
著者
著者名
Abe, Dsisuke
収録物名
Learner Corpus Studies in Asia and the World
巻(号)
2
ページ
383-393
出版者
神戸大学国際コミュニケーションセンター
刊行日
2014-05-31
公開日
2014-06-25
抄録(自由利用可)
This study combines the Stanford Parser, a syntactic analysis software, with the Nagoya Interlanguage Corpus of English, a learner corpus, in order to compare the phrase structures used by native speakers and learners of English. The use of syntactic analysis allows the extraction of complete phrase structures rather than the structurally incomplete word sequences which frequently appear when looking at multi-word expressions through the more traditional n-gram approach. Examining complete phrases is arguably more appropriate than studying all word sequences, as a past study has shown that both native and non-native speakers use phrases as their main units in writing. This study aims to complement a previous study which used parallel corpora by using non-parallel corpora. Although the use of parallel corpora is advantageous in that it provides controlled groups of texts similar in nature (e.g. topic distribution), there is also a limitation in that the words and structures used in one corpus is heavily affected by the other corpus. This paper examines the major phrase categories and the most frequent phrase structures appearing in the corpus. The results of this study supported previous findings that learners have less variation and more repetition in their multi-word expressions than native speakers.
カテゴリ
Learner Corpus Studies in Asia and the World
>
2号(2014-05-31)
紀要論文
詳細を表示
資源タイプ
departmental bulletin paper
言語
English (英語)
ISSN
2187-6746
OPACで所蔵を検索
CiNiiで学外所蔵を検索
関連情報
URI
http://www.solac.kobe-u.ac.jp/
ホームへ戻る